Search AI Products and News

Fudan University and Jieyue Xingchen, a leading domestic AI innovation company, recently announced the upcoming release of OmniSVG, an end-to-end multimodal SVG generation model. This news has quickly garnered widespread attention in the technology and design fields. According to AIbase, OmniSVG's core strength lies in its powerful generation capabilities, supporting vector graphic generation from simple icons to complex anime characters, providing a new intelligent solution for digital art creation. The launch of this model is poised to redefine the technical boundaries of vector graphic generation. Multimodal Generation: Flexible response.

Recently, research teams from Tsinghua University, Fudan University, and Stanford University jointly released an Agent development framework called 'Eko'. This framework aims to help developers quickly build production-ready 'virtual employees' using simple code and natural language. The Eko framework can take over a user's computer and browser, performing various tedious tasks on behalf of humans. With Eko, users can achieve automated data collection, testing, and file management, among other functionalities. For instance, users can set Eko to automatically gather information from Yahoo Finance.

On November 28, Fudan University officially released the "Guidelines for Using AI Tools in Undergraduate Theses (Draft)", which provides detailed regulations on the application of AI tools during the thesis writing process. This move marks the first instance of a Chinese university issuing dedicated regulatory management documents in this field, filling a policy gap. According to the guidelines, Fudan University allows students to use AI tools under specific conditions. Students may use these tools upon obtaining approval from their supervising teachers, provided that it does not hinder the assessment of their innovative capabilities.

Researchers from Fudan University and Baidu have jointly developed a new AI model called Hallo2, which can generate character animations in 4K resolution for durations of up to several hours, with precise control via voice and text prompts. Traditionally, producing high-quality character animations requires significant time and manpower. The introduction of Hallo2 is expected to completely change this situation, bringing revolutionary changes to fields such as filmmaking, virtual assistants, and game development.

The expert team formed by the MIFA Laboratory of the Shanghai Jiao Tong University Qingtuan Research Institute and the Endocrinology Department of Zhongshan Hospital affiliated with Fudan University has recently developed a dedicated large model for diabetes named Diabetica. This model integrates the language processing capabilities of large models with specialized knowledge in diabetes, aiming to provide intelligent support for doctors, patients, and medical education. The team employed a novel framework to establish a comprehensive data processing system, created high-quality datasets and evaluation benchmarks, and fine-tuned the large language model, significantly enhancing its performance on diabetes-related tasks.

With the rise of generative AI, the prompt trading market is rapidly expanding. However, current trading platforms, exemplified by PromptBase, are predominantly seller-driven in pricing and lack objective price measurement standards. In response to this challenge, the multimedia and intelligent security team at Fudan University has proposed an innovative prompt trading model aimed at better adapting to a future buyer-driven market. This new trading model consists of two main phases: prompt category selection and pricing strategy formulation. In the first phase, the platform employs a greedy search-based multi-armed bandit algorithm to...

A research team from Fudan University has developed the 'Concept Erasure Technique' (RECE), which can efficiently and accurately modify inappropriate ideas in AI models, significantly reducing the likelihood of generating inappropriate images while maintaining the AI's other creative capabilities. This technology uses a 'closed solution' to locate and modify specific parts, ensuring the AI behaves more appropriately during creation and stays on track. Experiments show that the RECE technology can effectively enhance the morality and quality of AI creations, providing new directions for the future development of AI, making future AI assistants smarter and more perceptive.

In the field of 3D digital humans, research teams from Nanjing University, Fudan University, and Huawei Noah's Ark Lab have proposed innovative solutions to address the issues of multi-view consistency and insufficient emotional expressiveness. They developed the EmoTalk3D dataset, which includes calibrated multi-view videos, emotion annotations, and frame-by-frame 3D geometric information. By constructing a mapping framework from 'speech to geometry to appearance', the research team introduced a new method that enables the synthesis of 3D talking avatars with controllable emotions, significantly enhancing lip synchronization and rendering quality.

Fudan University has introduced SpeechGPT, a large language model designed to understand and generate both speech and text. By discretizing speech signals, it enables compatibility with text modalities, allowing for emotional perception and multi-style speech generation based on context and instructions. The training strategy includes modal adaptation pre-training, cross-modal instruction fine-tuning, and modal chain instruction fine-tuning to en....

Hallo, an open source project by Fudan University that generates speaking videos based on audio and images, has now integrated with the ComfyUI plugin. Although the installation process requires a significant number of dependencies and has a relatively high barrier to entry, the emergence of this open source ecosystem offers more possibilities and enjoyment for subsequent transformation and other processes.

Fudan University Affiliated Zhongshan Hospital and Guangqi Huiyu have jointly released the Guangyu Medical Large Model, deeply participating in the customization of medical AI capability models that can be applied to various medical scenarios such as disease diagnosis and doctor-patient communication. The model's professionalism and credibility are ensured through high-quality medical training corpus. The Guangyu Medical Large Model, based on hundreds of billions of parameters and high-quality medical knowledge data, aims to improve the quality and efficiency of medical services and promote the comprehensive application of medical AI.

🧠 Research Background: The development roadmap of AGI is divided into five levels. LLMs have reached the second level, internet scale, but require perception and action space to reach the third and fourth levels. 🌟 Agent Composition: A combination of control end (LLMs), perception end, and action end for information processing, decision making, and environmental interaction. 📈 Application Scenarios: Three levels of single-agent tasks: daily task-oriented, innovation-driven, and lifecycle-oriented; multi-agent cooperative and adversarial interactions; human-machine interaction guidance - execution and partnership modes. 🚀 Agent Society: Individual behavior + collective behavior as three key components of personality.